Specify p-space configs in rs2

[Paolo Celani]

If you cannot optimize your state even with 90 configurations in the
p-space, I guess
the problem might lie somewhere else and I advice you to try the
following.
There is the possibility in RS2 to optimize any number of reference
states and
then select the ones for which you want to perform RS2 calculations.
This is obtained with variable nstati.
nstati is normally set to the maximum state specified in the STATE card
but you can
set it to any number by adding the following card to your input:

option,nstati=10

This will evaluate 10 reference wavefunctions (you said that when asking
for 10 roots
the RS2 code behaves correctly) and then you can specify the roots that
you want as
you did before.

Let me know if this works.

Bye

> 
> Hello,
> 
> Is it possible to specify that the rs2 module include certain p-space
> configurations (i.e., explicitly input the csf's)?  The molpro
> documentation for the CI program indicates this may be possible, but I
> don't understand how to do it.  Here is what the manual says under
> specifying the p-space configurations:
> 
>         Further configurations are added if either required by select or
> if configurations are found which are degenerate to the last p-space
> configuration. A minimum number of npspace is automatically determined
> from the state specifications.
>         How do I use select to define the p-space????  (it seems to be
> used to define the reference space)
> I have tried using npspace up to 90, but it is not finding the p-space
> configurations I need to converge the particular root.  The details are
> not important (I think) becuase I know that the root I want is there and
> I can converge it if I choose several roots in the rs2, but that leads
> to other complications -- so I want to tell the program to include in
> the p-space a few configurations which are prety high in energy, but are
> very important for the particular root I need (this is a highly excited,
> charge-transfer state, so there are many lower energy configurations in
> the reference space which essentially describe the charge distribution
> of the NON charge-transfer ground and excited states.
> 
> Thanks in advance for your input
> 
> Mike
> 
> Dr. Mike Falcetta
> 
> Associate Professor of Chemistry
> Roberts Wesleyan College
> 2301 Westside Drive
> Rochester, NY 14624
> 
>   ------------------------------------------------------------------------
>   ------------------------------------------------------------------------
> To add an item, send mail to molpro-user at tc.bham.ac.uk
> 
>   ------------------------------------------------------------------------